During the past two decades, there has been an increasing interest in the development of thermoacoustical cooling engines (pumps) for a variety of commercial, military and industrial applications. Interest in thermoacoustic cooling has accelerated rapidly with the production ban of chlorofluorocarbons (CFC's). Thermoacoustic refrigerators can be constructed such that they use only inert gases, which are non-toxic and do not contribute to ozone depletion, nor to global warming. Exemplary prior art designs for thermoacoustic engines and refrigerators are shown in the following patents: U.S. Pat. Nos. 4,398,398; 4,489,553, 4,722,201, 5,303,555, 5,647,216, 5,953,921, 6,032,464, and 6,314,740.
For a complete appreciation of the present invention, an understanding of earlier heat engines is beneficial.
Commercial Failure of the Stirling Cycle
The idea of passing a gaseous working fluid back and forth through a porous medium of high heat capacity (a “regenerator”) to improve the efficiency of a heat engine can be traced back to the invention in 1816, by Rev. Robert Stirling in England, of the thermodynamic cycle that bears his name. Although that invention was concerned with the production of useful mechanical work from heat, it was subsequently recognized that the Stirling cycle could be reversed to produce useful cooling, if mechanical energy was provided to the system.
The Stirling cycle has been attractive both as an engine and as a refrigerator for nearly two centuries because it could, in principle, achieve the maximum efficiency allowed within the constraints of the First and Second Laws of Thermodynamics. This limit of thermodynamically perfect performance is called the Carnot limit. Although an ideal Stirling engine or refrigerator could (in principle) exhibit Carnot performance, neither Stirling engines nor refrigerators ever achieved large-scale commercial success. A few engines based on the Stirling cycle have been used as the primary power source in submarines and many small refrigerators based on the Stirling cycle have been used to cool infrared detection electronics for military applications such as night vision goggles.
There are several reasons why this efficient approach to power production or refrigeration has not yet become commercially viable in most applications. The fundamental reason is that the improved efficiency (and more recently, the reduced environmental impact) of Stirling cycle devices was not an adequate incentive for its widespread adoption because the additional complexity and associated capital cost of the heat exchangers required by the Stirling cycle was not economically justified. In engine applications, the internal combustion engine was favored over the Stirling engine because it could exploit the high-temperature combustion of the fuel without requiring the solid parts of the engine to reach the same high temperature as the combustion products. After the energy was extracted from the combustion process, the excess heat carried by the combustion products could be exhausted directly to the atmosphere. No separate heat exchanger was required to exhaust waste heat from the engine, as required in closed cycle engines
In refrigeration applications, the vapor-compression (Rankine) cycle has been the dominant means for mechanical production of refrigeration. Although the Rankine cycle is less efficient than an ideal Stirling cycle, the additional mechanical complexity of a Stirling refrigerator and the cost of the heat exchangers needed for Stirling cycle refrigeration was, again, not economically justified. In a vapor-compression refrigerator, the vaporized working fluid could be used to extract the heat directly from the refrigeration load without requiring a secondary heat exchanger and a secondary heat exchange fluid. Because the phase-change of the working fluid exploited by the Rankine cycle was accompanied by a large latent heat, it was possible to produce vapor-compression refrigerators for cooling loads as small as a few tens of watts or as large as air conditioners with a cooling capacity equivalent to the energy absorbed by the melting of 2,000,000 pounds of ice per day (about 3.5 megawatts of useful cooling power).
Recent Developments
During the 20th century, many improvements to the Stirling cycle, for both refrigeration and for the conversion of heat to mechanical work, have been made. Thus far, none of these improvements have been sufficient to warrant the replacement of either the internal combustion engine or the vapor-compression refrigeration process by devices using a Stirling cycle. During the final quarter of the 20th century, an awareness of the environmental impact of both the internal combustion engine and the chlorofluorocarbons (CFCs) and other man-made chemicals used in most vapor-compression refrigerators and air conditioners became widespread. The global effects of stratospheric ozone depletion caused by CFCs, and the anthropogenic contributions to global warming produced by “greenhouse gases”, as well as other more localized effects such as “acid rain,” have stimulated a careful re-examination of both engine and refrigeration technology.
Beginning in the early 1980's, “thermoacoustics” has been one path that has been pursued to provide a new paradigm for production of environmentally friendly and energy-efficient alternatives to internal combustion engines and vapor-compression refrigerators. The thermoacoustic paradigm attempts to use the pressure oscillations and gas motions associated with sound waves to execute engine and refrigeration cycles with a minimum of mechanical moving parts. This is a conceptual break from the 19th century approach, in use to this day, that employs mechanical contrivances such as lubricated pistons moving in close-fitting cylinders, mechanically-actuated valves, flywheels, linkages, cams, etc., to impose the pressure changes and gas motions required to execute the cyclic processes that produce mechanical power or useful refrigeration. The first attempt to produce a practical “acoustical heat-pumping engine” (a thermoacoustic refrigerator) was patented by Wheatley, Swift, and Migliori in 1983 (see U.S. Pat. No. 4,398,398).
The Backhaus/Swift Engine
Since the invention of Wheatley, et al., there has been a continuous effort to produce thermoacoustic engines and refrigerators that would have the simplicity and robustness that came with the elimination of most mechanical parts, while achieving efficiencies that were comparable to or better than internal combustion engines and vapor-compression refrigerators. In 1999, Scott Backhaus and Greg Swift, both from Los Alamos National Laboratory in New Mexico, published the results of an experiment that used the thermoacoustic paradigm to produce a Stirling cycle engine that had a thermal efficiency of 30% [see “A thermoacoustic-Stirling heat engine,” Nature 399, 335–338 (1999)]. Their experimental device combined an acoustic phasing network and acoustic resonator to produce a one-horsepower Stirling cycle engine that was as efficient as a gas-powered automotive internal combustion engine but required no moving parts.
A version of the Backhaus/Swift engine that is suitable for refrigeration applications is shown in FIG. 1, which is taken from U.S. Pat. No. 6,032,464 (originally FIG. 6) patented by Swift, Backhaus & Gardner. The engine 1 included a driver or sound source 2 (in this case an intrinsically irreversible thermoacoustic engine) attached to a pressure vessel 3. The engine includes a toroidal path defined by an inertance tube 4, a secondary chamber (compliance) 5, and a thermal buffer tube 6. A flexible diaphragm 7 is attached to one end of the thermal buffer tubes 6 to act as a mass flux suppressor. The acoustic power circulates clockwise through the toroidal path, as indicated by arrows A. Thermal components 8 are provided in the toroidal path, and include a regenerator 10, and a first heat exchanger 9, and second heat exchanger 11. The inertance tube 4 and compliance 5 form an acoustical phasing network to produce pressure oscillations and gas flows through the engine's regenerator that are in phase, as required by the Stirling cycle. Further aspects of the Backhaus/Swift engine will be appreciated by those of skill in the art upon a complete review of U.S. Pat. No. 6,032,464 and the above-referenced article by the inventors. Unlike the original designs for a “traveling wave heat engine” patented by Peter Ceperley [see “Traveling wave heat engine,” U.S. Pat. No. 4,114,380 (Sep. 19, 1978); “Split mode traveling wave ring-resonator,” U.S. Pat. No. 4,686,407 (Aug. 11, 1987); and “Resonant traveling wave heat engine,” U.S. Pat. No. 4,355,517 (1982)], Swift, Backhaus & Gardner recognized that the ratio of the pressure to the volumetric velocity (the acoustic impedance) of the gas undergoing the Stirling cycle within the regenerator had to be much higher than the acoustic impedance that is characteristic of an acoustic traveling wave.
One drawback to the Swift, Backhaus & Gardner approach for either a thermoacoustic-Stirling engine or refrigerator is that their acoustic network creates a toroidal flow path that includes the regenerator and its associated heat exchangers. The toroidal flow path was also present in the designs of Ceperley. Such a flow path is shown in FIG. 1. The toroidal flow path allows steady circulation of the working fluid (Arrows A) driven by the flux of acoustic energy through the regenerator. That acoustically induced flow is known as Gedeon streaming and it can produce substantial reductions in the efficiency of the engine or refrigerator by convection of heat in the “wrong” direction.
For the Backhaus/Swift engine application, Swift, et al., invented a “jet pump” that used the asymmetry of inflow and outflow through a tapered orifice to produce a time-averaged backpressure to suppress the deleterious streaming flow. In the refrigeration application, they inserted a flexible diaphragm (7 in FIG. 1) to block the acoustically induced steady mass flow.
Acoustically induced streaming flow was deemed to be so detrimental for engine performance, the Los Alamos Thermoacoustics Group, headed by Dr. Swift, has recently abandoned the toroidal geometry for their next engine and has returned to a straight resonator. Their new design, which eliminates the toroidal flow path, is called “Cascade” and has been described in a recent publication [see S. Backhaus and G. Swift, “New varieties of thermoacoustic engines,” Proc. 9th Int. Congress on Sound and Vibration (July, 2002)].
In addition to the possibility of steady flow through the regenerator, another drawback to the thermoacoustic devices of Swift, et al. is the use of a column of gas to provide the inertance element (4 in FIG. 1), that forms a Helmholtz resonant “acoustic phasing network,” in conjunction with the gas stiffness that functions as compliance element 5. The oscillatory gas flow through the inertance element generates three types of hydrodynamic dissipation, which reduces the overall efficiency of the heat pumping process. At any amplitude, there are viscous boundary layer losses on the inner surface of inertance tube 4. As the amplitude increases, the oscillatory boundary layer becomes unstable and the flow becomes turbulent, further increasing power dissipation, which can be calculated from the correlations presented in the well know Moody diagram.
At the high amplitudes required for commercially acceptable volumetric power density, there are also exit-entrance losses at both ends of inertance tube 4. In long piping systems, these exit-entrance losses are known as “minor losses” or “head losses.” In thermoacoustic devices, such as that described by Swift, et al., above and by de Blok and Van Rijt below, these losses constitute a substantial fraction of the total input power. For the engine described by Backhaus and Swift, the thermoviscous losses in the inertance tube 4 is 4.9% of the total input power [see S. Backhaus and G. W. Swift, “A thermoacoustic-Stirling heat engine: Detailed study,” J. Acoust. Soc. Am. 107(6), 3148–3166 (2000)]. The “minor losses” at the inertance tube exit and entrance constituted a loss of 10% of the total input power.
The de Blok/Van Rijt Engine.
At about the same time as the invention of the Backhaus/Swift engine, C. M. (Kees) de Blok and N. A. H. J. Van Rijt, in the Netherlands, patented another version of a traveling-wave phased, impedance enhanced thermoacoustic-Stirling engine. One embodiment of this engine/refrigerator is shown in FIG. 2, which originally appeared in U.S. Pat. No. 6,314,740 (also as FIG. 2). This design includes a piston which is joined to a rigid enclosure by a flexible bellows. An electromechanical actuator 16 is attached to the piston-bellows combination 17, which is joined to the rigid enclosure 15. The rigid enclosure 15 contains the thermoacoustic elements of this refrigeration system. An acoustic phase control bypass 18 is formed by an internal connection tube 19. A cold heat exchanger is shown at 20, with a cold transport fluid inlet 20a and outlet 20b provided for connection to a refrigeration load. A hot heat exchanger is shown at 21, with hot transport fluid inlet 21a and outlet 21b providing a means to exhaust the waste heat that is pumped by the regenerator 22. While appearing physically dissimilar to the Swift, Backhaus & Gardner refrigerator, the de Blok/Van Rijt approach also introduces an effectively toroidal flow path. The specification does not address the detrimental consequences that such a path entails, or the losses that are produced by the “gas-filled bypass element” 18 that functions as the inertance element in the acoustical phasing network.
The TRITON Project
During the same time Swift, et al., and de Blok and Van Rijt were developing the thermoacoustic-Stirling devices discussed above, the Applied Research Laboratory at The Pennsylvania State University was funded by the US Navy, through the Office of Naval Research, to produce a larger version of the Shipboard Electronic Thermoacoustic Chiller (SETAC). The SETAC device is shown in FIG. 3, which is taken from U.S. Pat. No. 5,647,216 (originally FIG. 1). As shown, this thermoacoustic device is a double-ended device with drivers located at each end. Thermal components are located adjacent each driver, with the thermal components, including a stack, and a pair of heat exchangers at each end. The SETAC device was tested on board the USS Deyo (DD-989), a Spruance-class destroyer in the Atlantic Fleet, in April 1995. It demonstrated a maximum cooling capacity of 419 watts. The TRITON Project was an attempt to increase the cooling capacity of a SETAC-like device to 10 kilowatts; a cooling capacity equivalent to the latent heat absorbed by the melting of 3 tons (hence, TRI TON) of ice per day. (One ton of cooling is defined as 36,000 Btu/hr=3,517 watts.)
As part of the TRITON Project, flexure seals that employed metal bellows were developed that could function without fatigue failure at acoustic frequencies. These bellows were combined with mechanical springs and moving-magnet linear motors to produce electrodynamic loudspeakers that had power-handling capacities as large as 5 kW and electroacoustic conversion efficiencies that were nearly 90% (see U.S. Pat. No. 6,307,287). The measured efficiencies of these moving-magnet “loudspeakers” were found to be in excellent agreement with theoretical performance predictions of Wakeland [see R. S. Wakeland, “Use of electrodynamic drivers in thermoacoustic refrigerators,” J. Acoust. Soc. Am. 107(2), 827–832 (2000)], using measurements of the moving-magnet linear motor parameters that characterize the electromagnetic force factor and the electrical and mechanical dissipation.
The TRITON device also employed a double-Helmholtz resonator geometry that was similar to that used in SETAC. During testing of TRITON, the dissipation of acoustic power associated with the high-velocity oscillatory gas motion, particularly though the neck of the Helmholtz resonator and through the transitions between the neck and the two “bulbs,” led to unacceptably large nonlinear hydrodynamic losses. These hydrodynamic losses produced a substantial reduction in the overall performance of that refrigeration unit.
The Bellows Bounce Thermoacoustic Device
Two important lessons were learned from the TRITON Program, both in (i) the development of bellows to provide a low-loss, reliable, dynamic pressure flexure seal, and in (ii) the appreciation of the non-linear hydrodynamic losses associated with oscillatory gas flows at high Reynolds number and at transitions between parts of the resonator with different cross-sectional areas that is known as “minor loss” or “head loss.” It was recognized by the inventors that the resonator losses could be entirely eliminated and the size of a thermoacoustic chiller, for a given cooling capacity, could be substantial reduced, if the thermoacoustic core (regenerator and heat exchangers) and phasing network (inertance and compliance) were contained entirely within the bellows. Furthermore, it was recognized by the inventors that the resonant enhancement of the pressure oscillations created by the double-Helmholtz resonator could be duplicated, without the non-linear hydrodynamic losses inherent in the high-velocity gas motion through the neck, by using the elastic stiffness of the gas contained with the bellows and the moving mass of the linear motor and its attached piston, to create a mechanical resonator rather than a purely acoustic resonator used in TRITON, as well as all of the earlier thermoacoustic refrigerators. This novel resonator was named a “Bellows Bounce” compressor by its inventors, and is the subject of U.S. provisional patent application Ser. No. 60/372,008, filed Apr. 10, 2002, and a co-pending U.S. patent application entitled “Compliant Enclosure for Thermoacoustic Devices,” Ser. No. 10/409,855, filed Apr. 9, 2003, the entire contents of both of which are incorporated herein by reference.
An alternative to conventional bellows was also developed, and is the subject of U.S. provisional patent application Ser. No. 60/371,967, filed Apr. 10, 2002, and a co-pending U.S. patent application entitled “Cylindrical Spring with Integral Dynamic Gas Seal,” Ser. No. 10/409,760 filed Apr. 9, 2003, the entire contents of both of which are incorporated herein by reference. The cylindrical spring with integral dynamic gas seal provides an alternative to a more typical bellows that may enable greater design flexibility, lower production cost, and a significantly smaller surface area compared to a conventional bellows of equal volume with similar height and diameter.